The optimal fusion problem for the state estimation of discrete-time stochastic singular systems with multiple sensors and correlated measurement noise is considered. The key idea is to convert the original singular system into an equivalent group of nonsingular systems. Based on the state estimation for each local nonsingular system，the optimal full-order filters and smoothers are obtained for the original system using the optimal weighted fusion algorithms in the linear minimum variance sense. A simulation example shows that the fusion estimator is better than each local one.